Title: Message
Jon was right, there was a problem with the SNMP check.
Basicly what happens is that both Servers Alive and the control used for the SNMP check are checking if the timeout hasn't pasted yet and IF Servers Alive sees it before the control, the control can't fire it's timeout-event and this could give a false UP.
It's fixed in 3.3.1135.
Oh yeah 3.3.1135 was silently released today...I suppose telling it on this list makes it less silent?
 
 
 
 

dirk.

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of Kevin Stone
Sent: Sat Sep 21 4:42 PM
To: [EMAIL PROTECTED]
Subject: RE: [SA-list] SNMP checks

Since you are pulling SNMP values via IP why don't you make them dependent on a ping check?  The ping will let you know when the UPS is physically up or down and the SNMP check will provide details. 
 
You could also choose another counter that doesn't reall matter but returns a non-zero value.  When that value goes to zero you know the UPS is down and when it returns to non-zero you will know it's running.
 
-Kevin
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Dolinar, Jon
Sent: Friday, September 20, 2002 11:25 PM
To: [EMAIL PROTECTED]
Subject: [SA-list] SNMP checks

SA 3.2.1087
 
I have some SNMP checks (ups's)  that check to see the value is >0 . That works fine as long as the device is actually online and running I get the corect result.  If the device is off and it returns nothing the check passes as >0 and therfore running? So when the UPS goes on battery I get alerted but when the battery runs out the UPS says it is running again?

Reply via email to